There are several goals in the development of biological assays, including utilization of a minimal amount of assay components and sample, simplicity in operation and high throughput capability. Assays preferably require a minimal amount of assay components in order to minimize costs; this becomes a particular issue if certain assay components are expensive and/or a large number of assays are to be conducted. Ideally, assays require only a minimal amount of sample because often only a very limited amount of sample is available. The goal of simplicity of operation often means that the assay is preferably conducted in an integrated format in which all or most aspects of the assay can be conducted with a single device and minimal instrumentation. The goal of high throughput has become increasingly important in view of the trend in current research and drug discovery efforts to screen huge libraries of compounds to identify those that have a desired activity.
To address some of these problems, particularly the issue of minimizing the amount of sample and assay agents required to conduct an analysis, considerable effort has been invested in the development of microfluidic devices to conduct assays. These devices are characterized by using minute channels for the introduction, transport and mixing of the samples and agents necessary to conduct an assay. Unfortunately, current microfluidic devices suffer from a number of shortcomings that limit their usefulness. For example, current microfluidic devices often are manufactured from silicon chips with channels being etched into different silicon layers using established semi-conductor technologies. Such chips, however, are brittle and the stiffness of the material often necessitates high actuation forces. These forces and stresses can cause layers in a multilayer chip to separate from one another. The stiffness of the devices also imposes significant constraints on options for controlling solution flow through the microchannels.
Furthermore, solution flow is controlled at least in part through the use of electrodes to generate electric fields to move molecules and solution via electrophoresis and/or electroosmosis. Reliance on electrodes, however, creates several problems. One problem is that gas is often generated at the electrodes. This can increase pressure within the device potentially causing separation of microfabricated layers. The increased pressure and gas bubbles can also interfere with solution flow through the channels. Additionally, often an elaborate network of electrodes is required in order to achieve the desired level of control over solution transport. Fabrication of such a network can be complicated and increases the expense of the devices. The need for such networks also becomes particularly problematic if a device is to be prepared that includes a large number of branch channels to facilitate multiplexed and high throughput assay capabilities. Moreover, the use of electrical fields to control solution flow can be problematic for applications involving cells as application of the electric fields can negatively affect the cells, often killing them. Consequently, there remains a significant need for improved microfluidic devices, particularly those that are amendable to a wide range of cellular assays and that have high throughput capabilities.